Abstract PD5-10: Impact of immune markers on response to neoadjuvant de-escalated T-DM1 or trastuzumab with/or without endocrine therapy in HR+/HER2+ early breast cancer: A translational subproject of the WSG-ADAPT-HER2+/HR+ trial

Media type: E-Article

Title: Abstract PD5-10: Impact of immune markers on response to neoadjuvant de-escalated T-DM1 or trastuzumab with/or without endocrine therapy in HR+/HER2+ early breast cancer: A translational subproject of the WSG-ADAPT-HER2+/HR+ trial

Contributor: Biehl, Claudia; Kolberg-Liedtke, Cornelia; Gluz, Oleg; Feuerhake, Friedrich; Christgen, Matthias; Kates, Ronald; Grischke, Eva-Maria; Forstbauer, Helmut; Braun, Michael; Warm, Mathias; Hackmann, John; Uleer, Christoph; Aktas, Bahriye; Schumacher, Claudia; Kuemmel, Sherko; de Haas, Sanne L; Deurloo, Regula; Wuerstlein, Rachel; Nitz, Ulrike; Kreipe, Hans H.; Harbeck, Nadia

imprint: American Association for Cancer Research (AACR), 2020

Language: English

DOI: 10.1158/1538-7445.sabcs19-pd5-10

ISSN: 1538-7445; 0008-5472

Keywords: Cancer Research ; Oncology

Origination:

Footnote:

Description: Abstract Background: Among patients with highly heterogeneous HER2+ breast cancer (BC), pathological complete response (pCR) (ypT0/is/ypN0) is associated with improved prognosis. The prospective, phase-II neoadjuvant 3-arm WSG HER2+/HR+ trial (ADAPT-TP) showed pCR rates of about 41.0% after 4xT-DM1+/-ET and 15% after trastuzumab+ET. Pre-planned translational analysis of ADAPT-TP has revealed several biomarkers with impact on pCR, such as PIK3CA mutation status, HER2-enriched subtype (by PAM50), Bcl-2 and most notably immune biomarkers (e.g.CD8/tumor center). In particular, higher CD8 protein expression values either at baseline and/or after one cycle (3w) of neoadjuvant therapy were favorable for pCR. Moreover, CD8 and PDL1 dynamics were indicators of early therapy response, and dynamical response was itself favorable for pCR. The impact of immune biomarkers suggests that tumor infiltrating lymphocytes (TILs) could serve as predictive markers for de-escalated anti-HER2 therapy, in particular T-DM1. Methods: Semi-quantitative, triplicate TIL measurements were performed in tissue obtained at baseline (TIL-0) and at 3 weeks (TIL-3). At each timepoint, the median of the three measurements was used for analysis. CD8 and PD-L1 expression measurements were obtained by mRNA at baseline at both time points by immunohistochemistry (IHC). Associations of semiquantitative TIL-0/TIL-3 levels with pCR and with other clinical/pathological measurements were analyzed using logistic regression, rank correlations, t and chi-square statistics. Results: TIL-0 and TIL-3 were available in 355/375 and 301/375 randomized patients, respectively. Associations of TIL-0, TIL-3 and TIL-dynamics (increase from TIL-0 to TIL-3) with CD8 and PD-L1 (including dynamics) and with clinical factors (tumor size, nodal status, ER and PR status, grade, patient age, Ki67 and menopausal status) revealed several prominent relationships: Both TIL-0 and TIL-3 were at least moderately correlated with IHC measurements of CD8 (at center, not at invasive margin) and of PD-L1 at both timepoints. At cycle 2, the correlation of TILs and CD8 was strong, particularly in the T-DM1 arms (r=.7, p&lt;.001). CD8 by mRNA was moderately correlated with TIL-0. Correlations between clinical factors and TILs at both timepoints were weak. These associations with known factors influencing pCR suggest a possible direct association of TILs with pCR; logistic regression using standardized TIL-0 revealed a modest (standardized) odds ratio of about 1.4 for pCR for TIL-0 in the T-DM1 arms, though not in the whole cohort. In particular, patients receiving T-DM1 with TIL0≥40% (about 8%) had pCR of 70% vs. 40% pCR for others (p=.016). We observed a dynamic increase of TILs during T-DM1 treatment of about 10 percentage points. However, neither these dynamics nor TIL-3 were significantly associated with pCR (in contrast to CD8 protein expression at cycle 2). Interpretation: To the best of our best knowledge, these are the first prospective data regarding the predictive value of TILs and TIL dynamics in HR+/HER2+ BC treated with de-escalated anti-HER2 therapy. Very high TIL levels (&gt;40%) were associated with an excellent pCR rate of 70% in TDM1treated patients. However, CD8 expression and dynamics of CD8 were more strongly predictive for efficacy of de-escalated treatment than TILs – despite promising data on baseline and/or on-treatment TILs as markers for improved prognosis and higher sensitivity to (dual) anti-HER2 blockade (with or without chemotherapy) in unselected HER2+ BC. Further research on static and dynamic immune markers (according to HR+ and HR-/HER2+ subtypes) is needed prior to final conclusions regarding the clinical impact of TILs in early HER2+ BC. Citation Format: Claudia Biehl, Cornelia Kolberg-Liedtke, Oleg Gluz, Friedrich Feuerhake, Matthias Christgen, Ronald Kates, Eva-Maria Grischke, Helmut Forstbauer, Michael Braun, Mathias Warm, John Hackmann, Christoph Uleer, Bahriye Aktas, Claudia Schumacher, Sherko Kuemmel, Sanne L de Haas, Regula Deurloo, Rachel Wuerstlein, Ulrike Nitz, Hans H. Kreipe, Nadia Harbeck, West German Study Group. Impact of immune markers on response to neoadjuvant de-escalated T-DM1 or trastuzumab with/or without endocrine therapy in HR+/HER2+ early breast cancer: A translational subproject of the WSG-ADAPT-HER2+/HR+ trial [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr PD5-10.

Access State: Open Access

Search in field:

Recently searched for: